From: Hao Fu[³Å»¨] (justin_fu@xxxxxxxxxxxxxxx)
Date: Thu Mar 01 2001 - 20:21:31 GMT-3
Hello,
If u r using c6500 in CatOS 5.4.2 or later, I think u have better to use
another macro command, 'set port host' , it will diable ur port channel
negoitation and trunking negoitation, turn the port fast on...it's quite
suitable for the single host attached on it.
Justin Fu
> One of my customers had a problem only with W2K machines and DHCP. His
> NT4.0 and 98 machines didn't need port fast. Possibly W2K has less of a
> delay between loading the lan driver (and activating the link) and looking
> for a DHCP server? Or maybe they were just faster machines. Or maybe W2K
> has a shorter timeout for the DCHP lease request? Anyway, I've been using
> portfast on almost all workstation ports for the past few months.
>
> Thanks,
>
> Chuck Church
> CCNP, CCDP, MCNE, MCSE
> Sr. Network Engineer
> Magnacom Technologies
> 140 N. Rt. 303
> Valley Cottage, NY 10989
> 845-267-4000 x218
>
>
> -----Original Message-----
> From: Scott Morris [mailto:smorris@mentortech.com]
> Sent: Thursday, March 01, 2001 4:44 PM
> To: 'Chuck Church'; ''Ccielab' (E-mail)'; 'Cisco@Groupstudy. Com
> (E-mail)'
> Subject: RE: Portfast
>
>
> It's not specific to Windows 2000 machines... Any machine that needs DHCP
> and boots up with any speed (less than 50 seconds), or any machine running a
> novell client where it would try a GetNearestServer and find nothing....
>
> Scott
>
> -----Original Message-----
> From: nobody@groupstudy.com [mailto:nobody@groupstudy.com]On Behalf Of
> Chuck Church
> Sent: Thursday, March 01, 2001 4:22 PM
> To: 'Ccielab' (E-mail); Cisco@Groupstudy. Com (E-mail)
> Subject: RE: Portfast
>
>
> If this bdpu guard works as it supposed to, I'll definitely use it. Windows
> 2000 machines seem to need portfast for DHCP, and almost all Windows
> machines need it for IPX. I've always pointed out to the customer about
> NEVER connecting other layer 2 devices to the ports I configured portfast
> on. This is good insurance.
>
> Chuck Church
> CCNP, CCDP, MCNE, MCSE
> Sr. Network Engineer
> Magnacom Technologies
> 140 N. Rt. 303
> Valley Cottage, NY 10989
> 845-267-4000 x218
>
>
> -----Original Message-----
> From: Latimer, Keith [mailto:latimer@naptheon.com]
> Sent: Thursday, March 01, 2001 11:13 AM
> To: 'McCallum, Robert'; 'John Chang'; 'Ccielab' (E-mail);
> Cisco@Groupstudy. Com (E-mail)
> Subject: RE: Portfast
>
>
> Check out the new portfast bpdu guard feature. It can shut down ports that
> have portfast enabled when detecting bpdus on the line.
> Keith
>
> -----Original Message-----
> From: McCallum, Robert [mailto:Robert.McCallum@let-it-be-thus.com]
> Sent: Thursday, March 01, 2001 10:44 AM
> To: 'John Chang'; 'Ccielab' (E-mail); Cisco@Groupstudy. Com (E-mail)
> Subject: RE: Portfast
>
>
> No,
>
> The problem occurs if he creates a loop i.e. you have a main switch a cable
> from the main switch goes to user A. User A decides to connect a hub and a
> few terminals - Outcome fine. User B then says hey user A can you access
> those terminals and the main network. User A says yeah how do you want to
> connect? User A says yes and inadvertently patches his own pc and the
> original connection that was from him to the main switch outcome is now main
> switch has 2 connections to the minihub. NOW spanning tree goes oh my and
> recalculates - outcome 30 second outage for everyone on that vlan. Then the
> users go home, switch off their kit and go to the pub.
> Next day..... The mini hub is switched back on - because portfast is enabled
> the ports go whoosh straight into forwarding mode - result - spanning tree
> goes oh my!! and recalculates.
>
> Outcome ------ You and every other support member run about like loonies
> trying to find this fault which occurs only when the user decides to switch
> on his equipment.
>
> -----Original Message-----
> From: John Chang [mailto:johnec@umich.edu]
> Sent: 01 March 2001 15:34
> To: McCallum, Robert
> Subject: RE: Portfast
>
>
> Let me see if I got this correct. If he only connects one mini-hub or
> mini-switch it is OK to have portfast on on the main switch. If he then
> connects another mini-hub or mini-switch onto the first mini-hub or
> mini-switch than there will be a problem. But when you connect 2 mini-hubs
> aren't you just extending the amount of ports and in a sense there is only
> one virtual mini-hub?
>
> At 03:24 PM 3/1/2001 +0000, you wrote:
> >yes, but only if he then connects another link to another hub / switch and
> >causes a bridging loop.
> >
> >-----Original Message-----
> >From: John Chang [mailto:johnec@umich.edu]
> >Sent: 01 March 2001 15:08
> >To: cisco@groupstudy.com
> >Subject: Portfast
> >
> >
> >In the below website it says not to have portfast on if you connect
> >switches, hubs, or routers. I understand that point but what if a user
> >connected a mini-hub (Ex. Linksys EtherFast 8-Port 10/100 Desktop Hub)
> >or unmanaged mini-switch (Ex. Farallon NetLINE 10/100 switch) so that he
> >could connect multiple computers. Would this cause any problems? Thank
> >you!
> >
> >
> >http://www-1.cisco.com/warp/public/473/12.html
> >
> >Note: The portfast feature should never be used on switch ports that
> >connect to other switches, hubs, or routers. These connections may cause
> >physical loops
> >and it is very important that spanning tree go through the full
> >initialization procedure in these situations. A spanning tree loop can
> >bring your network down. If portfast
> >is turned on for a port that is part of a physical loop, it can cause a
> >window of time where packets could possibly be continuously forwarded (and
> >even multiply) in
> >such a way that the network cannot recover.
> >
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:29:17 GMT-3